Integration with other tools:

1
RAPID Representation and Analysis of
Probabilistic Intelligence Data
Overview
Architecture

• Organizations
• Carnegie Mellon University J. G. Carbonell (PI),
  E. Fink, and A. Gershman
• DYNAMiX Technologies D. Dietrich and G. Mani

Analystinterface
Approved plans forproactive data collection

• Purpose
• RAPID will help intelligence analysts to
  accomplish the following tasks
• Draw probabilistic conclusions from available
  intelligence, which may include uncertain,
  missing, and partially inaccurate data
• Identify critical uncertainties and potentially
  surprising developments
• Formulate and assess hypotheses based on the
  available data
• Develop strategies for proactive collection of
  additional intelligence to resolve uncertainties,
  based on the analysis of cost / benefit trade-offs

Value-addedreasoning tools
Uncertain situation assessmentand
data-collection planning
Hypotheses,conclusions, and data-collection
plans
Uncertainty calculus andproactive probe planning
Proactiveintelligencecollection
Excel extension for the analysis of uncertainty,
probes, and proactive data collection
Processing ofdata streams
The deliverables will include the executable and
source code, documentation, demos, and empirical
results.
Matching of queries andinferences against new
data
Newintelligence
Scalable assessment ofuncertain intelligence

• Functionality
• Representation of uncertainty Novel
  representation of massive uncertain data, which
  supports fast semi-automated reasoning
• Inferences from uncertain data Scalable
  inference mechanism for reasoning about uncertain
  intelligence
• Analysis of critical uncertainties Assessment of
  uncertain situations, evaluation of data utility,
  and identification of important missing data
• Proactive intelligence planning Evaluation of
  available probes and construction of optimized
  intelligence-collection plans

Generalintelligencecollection
Relational database of uncertaindata and
inference rules
External API
OTHER PAINT SYSTEMS
Product
Schedule

• Role in PAINT
• The RAPID tools are major part of the PAINT
  strategy module,
• responsible for the following functions within
  PAINT
• Representation and analysis of uncertainty in the
  leadership and pathway models, and in the
  expected probe behavior
• Planning of proactive data collection based on
  available probes
• Analysis of uncertain probing results, and
  evaluation of their impact on the validity of
  given hypotheses and future data-collection plans

Short term

• Integration with other tools
• We have integrated RAPID with Excel and Oracle,
  which allows the use of standard Excel and Oracle
  tools together with the RAPID tools
• We are extending the Excel interface for the work
  with uncertain data, thus enabling analysts to
  build task-specific spreadsheets for the analysis
  of uncertain intelligence
• We are working on advanced general-purpose API
  for the integration of RAPID with other
  off-the-shelf packages and with the PAINT
  architecture

Long term
All versions of RAPID will demonstrate all main
capabilities, with increasing functionality over
time.
Empirical evaluation We expect that RAPID will
provide significant advantage over available
off-the-shelf tools, such as standard
spreadsheets and database systems. We plan to
compare the productivity of analysts using the
overall PAINT (including RAPID) with that of
analysts who perform the same tasks using the
PAINT architecture where RAPID is replaced with
standard off-the-shelf tools. We will view RAPID
as success if it consistently outperforms
off-the-shelf tools, and analysts report the
overall positive experience of using it.